Enhancing AppAuthentix recommender systems using advanced machine learning techniques to identify genuine and...

Area of Science:

• Computer Science
• Cybersecurity
• Artificial Intelligence

Background:

• The rapid expansion of the smartphone app ecosystem has led to an increase in counterfeit and malicious applications.
• Existing security measures are insufficient to effectively distinguish between legitimate and harmful apps, posing risks to consumers and app vendors.
• There is an urgent need for advanced technological solutions to enhance app store security and user trust.

Purpose of the Study:

• To develop and evaluate a novel system for authenticating mobile applications and securing app stores.
• To address the growing threat of fraudulent and harmful apps in the digital marketplace.
• To improve customer confidence in mobile application platforms.

Main Methods:

• Utilized Convolutional Neural Networks (CNN) for image analysis of app data.
• Employed Natural Language Processing (NLP) for extracting features from app-related text.
• Integrated a novel algorithm, AppAuthentix Recommender, for robust app identification and authentication.

Main Results:

• The integrated system demonstrated high accuracy in identifying legitimate and counterfeit mobile applications.
• Achieved an impressive accuracy rate of 98.25% in estimating mobile app authenticity.
• The developed technology significantly enhances app store security and enables effective mobile app verification.

Conclusions:

• The study presents a groundbreaking approach to mobile app identification, crucial in the era of rapid app development.
• The combination of CNN, NLP, and the AppAuthentix Recommender algorithm substantially improves app store security.
• These advancements contribute to safer mobile app usage and increased consumer trust in digital marketplaces.